WaPo vs Upshot Senate Model

[GaussLaw]

http://www.washingtonpost.com/wp-tran/politics/election-lab-2014
http://www.nytimes.com/newsgraphics/2014/senate-model/


I prefer Upshot.  The WaPo model is decent but has 2 big misses IMO, North Carolina and Michigan, with a bias in reverse directions. 

[NHLiberal]

They are both garbage

[ElectionsGuy]

Washington Post projection is way off. McConnell apparently has a better chance of winning than Franken, according to that. It has little consistency with the polls or any other data we track here.

The Upshot model I think is pretty accurate, on the other hand.

[RogueBeaver]

Upshot quite good, WaPo's model must be on hard drugs.

[illegaloperation]

FiveThirtyEight has a more believable model: http://fivethirtyeight.com/features/fivethirtyeight-senate-forecast-toss-up-or-tilt-gop/

Grimes (KY) and Nunn (GA) probably have better shot than that though.

Also, NC and AK are probably tilt D and not toss-up. AR should be toss-up.

WaPo's model must be on hard drugs.

Yup. There's no way that Hagan (NC) has 90% chance or that Terri Lynn Land (MI) has 56% chance.

There are many other things that I didn't mention.

[IceSpear]

Upshot

[Miles]

I have less disagreements with the Upshot's.

[windjammer]

I have less disagreements with the Upshot's.

This,
But both of these models are terrible.
WaPO for Michigan (lol). Upshot because for them Nunn has more chances to win than Pryor (lol).

[Supersonic]

Michigan as a higher probability of going GOP than Arkansas? lol WaPo.